*OPEN STATA OUTPUT FILE LOG *


*log using "C:\Users\gk57526\Dropbox\Confirmation Dynamics Project (Jason Byers)\Confirmation Delay & Senate Committees\2023 Version\Fall 2024\Statistics\Output\Committee Delay.APPENDIX G RESULTS.smcl", replace

log using "/Users/jasonbyers/Dropbox/Jason Byers/Co-Authored Projects/Projects with George Krause/Krause Projects/Confirmation Dynamics Project/Confirmation Delay & Senate Committees/2023 Version/Fall 2024/Statistics/Output/Committee Delay.APPENDIX G RESULTS.smcl", replace

  
  
**** JSB UPDATED DATABASE: ADDING EXECUTIVE NOMINATION POSITIONS COVERED IN OSTRANDER DATABASE FROM MAY 2012 THROUGH DECEMBER 2020 AND UPDATING ALL OTHER DATA [SUMMER/FALL 2023]: /// 
*** ADDITIONAL VARIABLES ADDED IN MARCH 2024 IN RESPSONSE TO LSQ REFEREE REPORTS ****


**** "EXECUTIVE DEFERENCE OR LEGISLATIVE CONSTRAINT? COMMITTEE FOUNDATIONS OF CONFIRMATION DELAY FOR U.S. EXECUTIVE BRANCH APPOINTMENTS" [KRAUSE & BYERS] ****

 

   
*****************************************************************************************************************************************************************************************
*****************************************************************************************************************************************************************************************
*****************************************************************************************************************************************************************************************
*****************************************************************************************************************************************************************************************
*****************************************************************************************************************************************************************************************
*****************************************************************************************************************************************************************************************
*****************************************************************************************************************************************************************************************
*****************************************************************************************************************************************************************************************
   
   
   



**** APPENDIX G SENSITIVITY ANALYSES: EVALUATING DIFFERENCES IN REPORTED MANUSCRIPT MODELS 1-4 BASED ON WHETHER EXECUTIVE NOMINEE WAS RECENTLY SENATE CONFIRMED IN PRIOR TWO CONGRESSES [PRIORCONFIRM==1] OR HAD NOT BEEN [PRIORCONFIRM==0] ****





**** RATIONALE: THOSE WHO WERE NOT PRIOR SENATE CONFIRMED SHOULD FACE LONGER COMMITTEE CONFIRMATION DELAY THAN THOSE WHO WERE CONFIRMED WITHIN THE PRIOR TWO CONGRESSES *** 






**************************************************************************************************************************************************************************************** **************************************************************************************************************************************************************************************** *************************************************************************************************************************************************************************************** ****************************************************************************************************************************************************************************************
			  			  
	
  
*** COMMITTEE CONTROL COVARIATES:       experience_median  / chair_experience_1; ///
***										ln_combills_workload; committeestaffsize [# Committee Staff]   
    
*** OSTRANDER CONTROL COVARIATES:       sendivide  polarization pres_app_m first90 preselection lameduck workload [Executive Civilian Nominations: Senate] 
***										female priorconfirm _Itier_2 _Itier_3 _Itier_4 defense infrastructure social


*** ADDITIONAL CONTROL COVARIATES:      pressenfloorabsdist [-] [|Senate Floor Median - President|]; kv_workload (# civilian executive nominations made in a given year/session)	
***										denied [-] [nominee previously denied in same Congress]; fvra [+] [= 1 if subject to FVRA 1998, = 0 otherwise]; 
***										firstrecess[-] [= 1 is nominated in July or August affected by August Recess, = 0 otherwise]; 
*** 									secondrecess [-] [= 1 is nominated in November or December affected by December Recess, = 0 otherwise]; major policy agency binary indicator [-];                      
***                                     committee-level & pressidential administration unit/fixed effects.
                                        
                                
							                       
 
	   
	   
	   

  
 * OPEN UPDATED "CONFIRMATION DELAY & SENATE COMMITTEES PROJECT" MANUSCRIPT DATABASE [12-23-2024] *
 
*use "C:\Users\gk57526\Dropbox\Confirmation Dynamics Project (Jason Byers)\Confirmation Delay & Senate Committees\2023 Version\Fall 2024\Statistics\Data\Committee Delay.MANUSCRIPT RESULTS.12-23-2024.dta", replace

use "/Users/jasonbyers/Dropbox/Jason Byers/Co-Authored Projects/Projects with George Krause/Krause Projects/Confirmation Dynamics Project/Confirmation Delay & Senate Committees/2023 Version/Fall 2024/Statistics/Data/Committee Delay.MANUSCRIPT RESULTS.12-23-2024.dta", replace




*********************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************




*** EVALUATE PSCD HYPOTHESIS FOR MODELS PRESENTED IN MANUSCRIPT: DISTINCTION BETWEEN PRIOR CONFIRMATION (PRIORCONFIRM==1) VERSUS OTHERWISE (PRIORCONFIRM==0) ***






*** MODEL G.1A: FULL SAMPLE: |COMMITTEE MEDIAN - PRESIDENT| & PRIORCONFIRM==0 [COX SEMIPARAMETRIC MODEL] ***

stcox  c.committee_pres1##i.sendivide  pressenfloorabsdist   experience_median  committeestaffsize ln_combills_workload   pres_app_m first90 preselection lameduck   kv_workload  polarization   workload  female priorconfirm denied  x_itier_2 x_itier_3 x_itier_4 defense infrastructure social fvra firstrecess secondrecess policy_majagency  i.kbcom_1  i.presrev  if priorconfirm==0,  vce(cluster kbcom_1)
*

* DESCRIPTIVE STATISTICS FOR EACH PARTISAN CONTROL REGIME [NOTE: THESE VARY ACROSS SUBSAMPLES OF INTEREST] *
sum wSenComm_committee_pres1 if e(sample) & sendivide==0, detail
sum wSenComm_committee_pres1 if e(sample) & sendivide==1, detail



** CONDITIONAL COEFFICIENT ANALYSIS TESTS: DIRECTION [+] ** 

* DIFFERENCE BETWEEN DIVIDED AND UNIFIED PARTISAN CONTROL OF SENATE & PRESIDENCY: INTERQUARTILE UNIT CHANGE IN "wSenComm_committee_pres1"  *
lincomest (committee_pres1 * 0.3083121 +  1.sendivide#c.committee_pres1 * 0.2583745) - committee_pres1 * 0.3083121, eform(hr)
matrix model1a = r(table)
mat list model1a

*
*
*
*



*** MODEL G.1B: FULL SAMPLE: |COMMITTEE MEDIAN - PRESIDENT| & PRIORCONFIRM==1 [COX SEMIPARAMETRIC MODEL] ***

stcox  c.committee_pres1##i.sendivide  pressenfloorabsdist   chair_experience_1  committeestaffsize ln_combills_workload   pres_app_m first90 preselection lameduck   kv_workload  polarization   workload  female priorconfirm denied  x_itier_2 x_itier_3 x_itier_4 defense infrastructure social fvra firstrecess secondrecess policy_majagency   i.kbcom_1  i.presrev  if priorconfirm==1,  vce(cluster kbcom_1)
*

* DESCRIPTIVE STATISTICS FOR EACH PARTISAN CONTROL REGIME [NOTE: THESE VARY ACROSS SUBSAMPLES OF INTEREST] *
sum wSenComm_committee_pres1 if e(sample) & sendivide==0, detail
sum wSenComm_committee_pres1 if e(sample) & sendivide==1, detail



** CONDITIONAL COEFFICIENT ANALYSIS TESTS: DIRECTION [+] ** 

* DIFFERENCE BETWEEN DIVIDED AND UNIFIED PARTISAN CONTROL OF SENATE & PRESIDENCY: INTERQUARTILE UNIT CHANGE IN "wSenComm_committee_pres1"  *
lincomest (committee_pres1 * 0.437 +  1.sendivide#c.committee_pres1 * 0.2466255) - committee_pres1 * 0.437, eform(hr)
matrix model1b = r(table)
mat list model1b



******************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************






*** MODEL G.2A: FULL SAMPLE: |COMMITTEE CHAIR - PRESIDENT| & PRIORCONFIRM==0 [COX SEMIPARAMETRIC MODEL] ***

stcox  c.chair_pres1##i.sendivide   pressenfloorabsdist   experience_median  committeestaffsize ln_combills_workload   pres_app_m first90 preselection lameduck   kv_workload  polarization   workload  female priorconfirm denied  x_itier_2 x_itier_3 x_itier_4 defense infrastructure social fvra firstrecess secondrecess policy_majagency   i.kbcom_1  i.presrev  if priorconfirm==0,  vce(cluster kbcom_1)
*


* DESCRIPTIVE STATISTICS FOR EACH PARTISAN CONTROL REGIME [NOTE: THESE VARY ACROSS SUBSAMPLES OF INTEREST] *
sum wSenComm_chair_pres1 if e(sample) & sendivide==0, detail
sum wSenComm_chair_pres1 if e(sample) & sendivide==1, detail



** CONDITIONAL COEFFICIENT ANALYSIS TESTS: DIRECTION [+] ** 

* DIFFERENCE BETWEEN DIVIDED AND UNIFIED PARTISAN CONTROL OF SENATE & PRESIDENCY: INTERQUARTILE UNIT CHANGE IN "wSenComm_chair_pres1"  *
lincomest (chair_pres1 * 0.2355347 +  1.sendivide#c.chair_pres1 * 0.2306869) - chair_pres1 * 0.2355347, eform(hr)
matrix model2a = r(table)
mat list model2a

*
*
*
*



*** MODEL G.2B: FULL SAMPLE: |COMMITTEE CHAIR - PRESIDENT| & PRIORCONFIRM==1 [CCOX SEMIPARAMETRIC MODEL] ***

stcox  c.chair_pres1##i.sendivide  pressenfloorabsdist   chair_experience_1  committeestaffsize ln_combills_workload   pres_app_m first90 preselection lameduck   kv_workload  polarization   workload  female priorconfirm denied  x_itier_2 x_itier_3 x_itier_4 defense infrastructure social fvra firstrecess secondrecess policy_majagency  i.kbcom_1  i.presrev  if priorconfirm==1,  vce(cluster kbcom_1)
*

* DESCRIPTIVE STATISTICS FOR EACH PARTISAN CONTROL REGIME [NOTE: THESE VARY ACROSS SUBSAMPLES OF INTEREST] *
sum wSenComm_chair_pres1 if e(sample) & sendivide==0, detail
sum wSenComm_chair_pres1 if e(sample) & sendivide==1, detail


** CONDITIONAL COEFFICIENT ANALYSIS TESTS: DIRECTION [+] ** 

* DIFFERENCE BETWEEN DIVIDED AND UNIFIED PARTISAN CONTROL OF SENATE & PRESIDENCY: INTERQUARTILE UNIT CHANGE IN "wSenComm_chair_pres1"  *
lincomest (chair_pres1 * 0.2730916 +  1.sendivide#c.chair_pres1 * 0.281416) - chair_pres1 * 0.2730916, eform(hr)
matrix model2b = r(table)
mat list model2b




*****************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************






*** MODEL H.3A: FULL SAMPLE: |COMMITTEE MEDIAN - PRESIDENT| & PRIORCONFIRM==0 [WEIBULL PARAMETRIC MODEL] ***

streg  c.committee_pres1##i.sendivide    pressenfloorabsdist   experience_median  committeestaffsize ln_combills_workload   pres_app_m first90 preselection lameduck   kv_workload  polarization   workload  female priorconfirm denied  x_itier_2 x_itier_3 x_itier_4 defense infrastructure social fvra firstrecess secondrecess policy_majagency   i.kbcom_1  i.presrev if priorconfirm==0,  distribution(weibull) vce(cluster kbcom_1)
*

* DESCRIPTIVE STATISTICS FOR EACH PARTISAN CONTROL REGIME [NOTE: THESE VARY ACROSS SUBSAMPLES OF INTEREST] *
sum wSenComm_committee_pres1 if e(sample) & sendivide==0, detail
sum wSenComm_committee_pres1 if e(sample) & sendivide==1, detail



** CONDITIONAL COEFFICIENT ANALYSIS TESTS: DIRECTION [+] ** 

* DIFFERENCE BETWEEN DIVIDED AND UNIFIED PARTISAN CONTROL OF SENATE & PRESIDENCY: INTERQUARTILE UNIT CHANGE IN "wSenComm_committee_pres1"  *
lincomest (committee_pres1 * 0.3083121 +  1.sendivide#c.committee_pres1 * 0.2583745) - committee_pres1 * 0.3083121, eform(hr)
matrix model3a = r(table)
mat list model3a

*
*
*
*



*** MODEL G.3B: FULL SAMPLE: |COMMITTEE MEDIAN - PRESIDENT| & PRIORCONFIRM==1 [WEIBULL PARAMETRIC MODEL] ***

streg  c.committee_pres1##i.sendivide    pressenfloorabsdist   experience_median  committeestaffsize ln_combills_workload   pres_app_m first90 preselection lameduck   kv_workload  polarization   workload  female priorconfirm denied  x_itier_2 x_itier_3 x_itier_4 defense infrastructure social fvra firstrecess secondrecess policy_majagency   i.kbcom_1  i.presrev  if priorconfirm==1, distribution(weibull) vce(cluster kbcom_1)
*

* DESCRIPTIVE STATISTICS FOR EACH PARTISAN CONTROL REGIME [NOTE: THESE VARY ACROSS SUBSAMPLES OF INTEREST] *
sum wSenComm_committee_pres1 if e(sample) & sendivide==0, detail
sum wSenComm_committee_pres1 if e(sample) & sendivide==1, detail



** CONDITIONAL COEFFICIENT ANALYSIS TESTS: DIRECTION [+] ** 

* DIFFERENCE BETWEEN DIVIDED AND UNIFIED PARTISAN CONTROL OF SENATE & PRESIDENCY: INTERQUARTILE UNIT CHANGE IN "wSenComm_committee_pres1"  *
lincomest (committee_pres1 * 0.437 +  1.sendivide#c.committee_pres1 * 0.2466255) - committee_pres1 * 0.437, eform(hr)
matrix model3b = r(table)
mat list model3b



******************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************




*** MODEL G.4A: FULL SAMPLE: |COMMITTEE CHAIR - PRESIDENT| & PRIORCONFIRM==0 [WEIBULL PARAMETRIC MODEL] ***

stcox  c.chair_pres1##i.sendivide   pressenfloorabsdist   chair_experience_1  committeestaffsize ln_combills_workload   pres_app_m first90 preselection lameduck   kv_workload  polarization   workload  female priorconfirm denied  x_itier_2 x_itier_3 x_itier_4 defense infrastructure social fvra firstrecess secondrecess policy_majagency    i.kbcom_1  i.presrev if priorconfirm==0, vce(cluster kbcom_1)
*

* DESCRIPTIVE STATISTICS FOR EACH PARTISAN CONTROL REGIME [NOTE: THESE VARY ACROSS SUBSAMPLES OF INTEREST] *
sum wSenComm_chair_pres1 if e(sample) & sendivide==0, detail
sum wSenComm_chair_pres1 if e(sample) & sendivide==1, detail



** CONDITIONAL COEFFICIENT ANALYSIS TESTS: DIRECTION [+] ** 

* DIFFERENCE BETWEEN DIVIDED AND UNIFIED PARTISAN CONTROL OF SENATE & PRESIDENCY: INTERQUARTILE UNIT CHANGE IN "wSenComm_chair_pres1"  *
lincomest (chair_pres1 * 0.2355347 +  1.sendivide#c.chair_pres1 * 0.2306869) - chair_pres1 * 0.2355347, eform(hr)
matrix model4a = r(table)
mat list model4a

*
*
*
*




*** MODEL G.4B: FULL SAMPLE: |COMMITTEE CHAIR - PRESIDENT| & PRIORCONFIRM==1 [WEIBULL PARAMETRIC MODEL] ***

streg  c.chair_pres1##i.sendivide   pressenfloorabsdist   chair_experience_1  committeestaffsize ln_combills_workload   pres_app_m first90 preselection lameduck   kv_workload  polarization   workload  female priorconfirm denied  x_itier_2 x_itier_3 x_itier_4 defense infrastructure social fvra firstrecess secondrecess policy_majagency      i.kbcom_1  i.presrev if priorconfirm==1,  distribution(weibull) vce(cluster kbcom_1)
*

* DESCRIPTIVE STATISTICS FOR EACH PARTISAN CONTROL REGIME [NOTE: THESE VARY ACROSS SUBSAMPLES OF INTEREST] *
sum wSenComm_chair_pres1 if e(sample) & sendivide==0, detail
sum wSenComm_chair_pres1 if e(sample) & sendivide==1, detail


** CONDITIONAL COEFFICIENT ANALYSIS TESTS: DIRECTION [+] ** 

* DIFFERENCE BETWEEN DIVIDED AND UNIFIED PARTISAN CONTROL OF SENATE & PRESIDENCY: INTERQUARTILE UNIT CHANGE IN "wSenComm_chair_pres1"  *
lincomest (chair_pres1 * 0.2730916 +  1.sendivide#c.chair_pres1 * 0.281416) - chair_pres1 * 0.2730916, eform(hr)
matrix model4b = r(table)
mat list model4b





*** CREATE FIGURE G1 BASED ON MODELS 1A/1B/2A/2B/3A/3B/4A/4B ****

**** FIGURE G1 ****

matrix A = J(8, 3, .)
matrix coln A = Point ll95 ul95
matrix rown A = 1 2 3 4 5 6 7 8

matrix A[1,1] = model1a[1,1]
matrix A[1,2] = model1a[5,1]
matrix A[1,3] = model1a[6,1]

matrix A[2,1] = model1b[1,1]
matrix A[2,2] = model1b[5,1]
matrix A[2,3] = model1b[6,1]

matrix A[3,1] = model2a[1,1]
matrix A[3,2] = model2a[5,1]
matrix A[3,3] = model2a[6,1]

matrix A[4,1] = model2b[1,1]
matrix A[4,2] = model2b[5,1]
matrix A[4,3] = model2b[6,1]

matrix A[5,1] = model3a[1,1]
matrix A[5,2] = model3a[5,1]
matrix A[5,3] = model3a[6,1]

matrix A[6,1] = model3b[1,1]
matrix A[6,2] = model3b[5,1]
matrix A[6,3] = model3b[6,1]

matrix A[7,1] = model4a[1,1]
matrix A[7,2] = model4a[5,1]
matrix A[7,3] = model4a[6,1]

matrix A[8,1] = model4b[1,1]
matrix A[8,2] = model4b[5,1]
matrix A[8,3] = model4b[6,1]



coefplot (matrix(A[,1]), ci((2 3))), grid(none) xline(1, lcolor(red%40) lpattern(dash)) xtitle("Hazard Ratio", size(small) margin(t=2)) ylabel(1 "Model G1.A" 2 "Model G1.B" 3 "Model G2.A" 4 "Model G2.B" 5 "Model G3.A" 6 "Model G3.B" 7 "Model G4.A" 8 "Model G4.B", labsize(small) noticks) mlabel format(%9.3f) mlabposition(12) mlabsize(vsmall) xlabel(0(1)3, angle(0) labsize(small) format(%9.1f)) msymbol(o) mcolor(black) msize(vsmall) title("FIGURE G1", size(med)) ciopts(lcolor(black)) legend(off) subtitle("Differential Partisan Control Effects of Committee-President Ideological Distance" "(Distinctions Between Nominees With and Without Prior Senate Confirmation)", size(small))

*graph save "Graph" "C:\Users\gk57526\Dropbox\Confirmation Dynamics Project (Jason Byers)\Confirmation Delay & Senate Committees\2023 Version\Fall 2024\Statistics\Graphics\FigureH1.gph", replace

graph save "Graph" "/Users/jasonbyers/Dropbox/Jason Byers/Co-Authored Projects/Projects with George Krause/Krause Projects/Confirmation Dynamics Project/Confirmation Delay & Senate Committees/2023 Version/Fall 2024/Statistics/Graphics/Appendix G/FigureG1.gph", replace









********************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************************


log close 
